[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20190304231426.GA6191@amd>
Date: Tue, 5 Mar 2019 00:14:26 +0100
From: Pavel Machek <pavel@....cz>
To: "Theodore Y. Ts'o" <tytso@....edu>
Cc: adilger.kernel@...ger.ca, jack@...e.cz,
linux-fsdevel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: 5.0.0-rc8-next-20190301+: kernel bug at fs/inode.c:513
Hi!
> (Adding linux-mm and moving linux-ext4 and linux-kernel to the bcc
> list...)
Umm. So I ... bcc them too?
> On Mon, Mar 04, 2019 at 05:02:55PM +0100, Pavel Machek wrote:
> >
> > This happened on trying to sync filesystems with unison:
> >
> > Any ideas?
> >
> > Should I be forcing fsck soon?
> > Pavel
> >
> > [12717.827444] ------------[ cut here ]------------
> > [12717.827465] kernel BUG at fs/inode.c:513!
>
> It seems unlikely that fsck is going to help. The BUG_ON in question
> appears to be:
>
> BUG_ON(inode->i_data.nrexceptional);
>
> This is an in-memory accounting variable which seems to be related to
> tracking DAX and shadow page entries --- and it seems very unlikely to
> be the sort of thing triggered by an on-disk corruption.
>
> Do you have a reliable repro for this? If so, the next step would be
> to bisect...
It happened just once... so I don't yet know.
I'm not even sure what file was affected. unison was showing some kind
of png from openstreetmap cache, but I could read it using md5sum just
fine.
-next is normally pretty boring, but it seems to get pretty
interesting around -final release...
Thanks,
Pavel
--
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)
Powered by blists - more mailing lists